Thanks for your reply, Gordon.
I just re-installed RH6.1 on a test system, this time without RAIDing
/boot, and everything else I did the same as before. Also, I did let the
mirror complete its initialization before restarting the system just after
the install (which is a first - but I don't know that it matters). This
time, I do not get the shutdown error [FAILED] described before.
*However*, I *still* get errors in dmesg logs... this time saying:
request_module[md-personality-3]: Root fs not mounted
do_md_run() returned -22
(note the number change) however, the system still boots, and performs as I
suspect it should... and the HDD LED indicators appear to me to actually be
working as a mirror, so I believe the mirroring to be working.
*However*, kernel upgrades *still* bomb. It's still a kernel panic, and
I'm beginning to believe that I'm doing something wrong with the kernel
upgrade (probably not doing something unique to md systems as I've done
other kernel upgrades without a hitch from source). This is what the
screen reads when it gets the kernel panic (which has varied a small bit,
but not much, from the previous installation when /boot was RAIDed):
...
Partition check:
hda: hda1 hda2 <hda5 hda6>
hdc: hdc1 hdc2 <hdc5>
RAMDISK: Compressed image found at block 0
VFS: mounted root (ext2 filesystem)
Loading raid0 module
raid0 personality registered
Loading raid1 module
raid1 personality registered
Oops! md1 not running, giving up!
Bad md_map in ll_rw_block
EXT2-fs: unable to read superblock
Oops! md1 not running, giving up!
Bad md_map in ll_rw_block
FAT bread failed
Oops! md1 not running, giving up!
Bad md_map in ll_rw_block
FAT bread failed
Oops! md1 not running, giving up!
Bad md_map in ll_rw_block
EXT2-fs: unable to read superblock
isofs_read_super: bread failed, dev=09:01, iso_blknum16, block 32
Kernel panic: VFS: Unable to mount root fs on 09:01
Because I'm at an "upgrade" stage, and not an "install" stage, the FAQs and
whatnot on linuxdoc.org don't really help me too much. They don't mention
these errors. Where would I have read (if I had read anything on linux
RAID beforehand) about not RAIDing /boot ? Why does the RedHat installer
allow me to do that if I shouldn't? And, why does it mostly work if it
shouldn't?
Thanks.
Lee Howard
At 11:55 AM 7/21/00 -0700, you wrote:
>On Fri, 21 Jul 2000, Lee Howard wrote:
>
>> Is anybody else with software RAID on their 6.x systems seeing the
>> following in their /var/log/dmesg logs? (It also appears in messages log.)
>>
>> request_module[md-personality-2]: Root fs not mounted
>> do_md_run() returned -22
>
>I'm not. :)
>
>> Turning off md devices
>> md2 still mounted device or resource busy [FAILED]
>
>Try shutting down to runlevel 1, rather than 6. Use lsof or fuser to
>determine what programs have files open on the root fs. It sounds like
>something isn't shutting down when init tells it to shut down.
>
>> My biggest problem is that when I try to upgrade the kernel to 2.2.16,
>> whether it be from source or by RPM the end result is *always* fatal when
>...
>> md1 = hda5 (32MB) & hdc5 (32MB) in
>> RAID1 for /boot
>
>AFAIK, that's not OK, and is probably the reason you can't upgrade your
>kernel. Try unmounting /boot, mounting /dev/hda5 on boot, and upgrading
>your kernel. If it works, don't make /boot a RAID device anymore.
>
>MSG
--
To unsubscribe: mail [EMAIL PROTECTED] with "unsubscribe"
as the Subject.
